www.nature.

com/scientificreports

OPEN PMMA-Etching-Free Transfer

of Wafer-scale Chemical Vapor
Deposition Two-dimensional
received: 10 May 2016
accepted: 19 August 2016 Atomic Crystal by a Water Soluble
Polyvinyl Alcohol Polymer Method
Published: 12 September 2016

HuynhVan Ngoc1, YongtengQian1, SukKilHan2 & DaeJoonKang1

We have explored a facile technique to transfer large area 2-Dimensional (2D) materials grown by
chemical vapor deposition method onto various substrates by adding a water-soluble Polyvinyl Alcohol
(PVA) layer between the polymethyl-methacrylate (PMMA) and the 2D material film. This technique
not only allows the effective transfer to an arbitrary target substrate with a high degree of freedom, but
also avoids PMMA etching thereby maintaining the high quality of the transferred 2D materials with
minimum contamination. We applied this method to transfer various 2D materials grown on different
rigid substrates of general interest, such as graphene on copper foil, h-BN on platinum and MoS2 on
SiO2/Si. This facile transfer technique has great potential for future research towards the application of
2D materials in high performance optical, mechanical and electronic devices.

Two-dimensional (2D) materials have attracted much attention due to their unique properties and great potential
in various applications. Controllable synthesis of 2D materials with high quality and high efficiency is essential
for their large scale device applications. Chemical vapor deposition (CVD) has been one of the most reliable
technique for the synthesis of 2D materials. To explore novel applications and the discovery of new phenomena in
these materials, it is necessary to develop a process to transfer high quality and large area 2D materials onto desir-
able substrates with great efficiency and high yield. Currently, the most commonly used transfer method relies on
the use of sacrificial polymethyl-methacrylate (PMMA) film to support the 2D layers and to prevent them from
folding during the etching of the growth substrate. However, the removal of PMMA is usually incomplete and
leads to surface contamination due to undissolved PMMA residues17. These contaminants are trapped on the
surface of the transferred 2D materials, resulting in the degradation of the intrinsic properties, and reduce the
reliability of devices fabricated from them. This effect also has consequences for the formation of multiple layer
heterostructures from 2D materials in that it severely limits future work on these, which are usually formed using
layer-by-layer transfer methods. Other transfer methods which do not involve PMMA have been also investi-
gated, including using thermal release tapes as transfer membranes1, direct transfer onto polydimethylsiloxane
(PDMS)2, or using two-layer structures consisting of polyethylene terephthalate (PET) and silicone3, and hot
pressing4. But so far, the wide applicability of these methods is still limited due to limitations in size, uniformity
and quality of the transferred 2D layers. Moreover, eliminating the metal etching step and transferring graphene
by a direct delamination from the growth substrate is the most effective way to avoid contamination from both
metal and PMMA impurities. Several direct delamination and transfer methods without PMMA have been
also proposed, such as, use of mechanical fracture testing instrument5,6 and Polyvinyl Alcohol (PVA) assisted
transferring7. However, the degree of freedom and uniformity in the transfer process is still significantly limited
by graphene not being in an isolated state, since the delamination and the transfer of graphene are performed
simultaneously using an adhesion layer. Moreover, wrinkling of the graphene surface due to the Poisson effect
of the elastomer stamp during the transfer process cannot be avoided. Especially, direct delamination methods

1
Department of Physics and Interdisciplinary Course of Physics and Chemistry, Sungkyunkwan University, 2066,
Seobu-Ro, Jangan-Gu, Suwon, Gyeonggi-do 16419, Republic of Korea. 2Teraleader Inc., 55-8, Techno-11ro,
Yuseong-gu, Daejeon 34036, Republic of Korea. Correspondence and requests for materials should be addressed to
D.J.K. (email: djkang@skku.edu)

Scientific Reports | 6:33096 | DOI: 10.1038/srep33096 1

www.nature.com/scientificreports/

Figure 1. Schematic illustration of PVA assisted transfer of CVD-grown 2D materials. (a) Graphene
transfer by copper etching method, (b) Single layer h-BN transfer by bubbling method.

to transfer 2D transition metal dichalcogenide materials grown on oxide substrates have not been demonstrated
due to their strong Van der Waals interaction between the 2D transition metal dichalcogenide materials and its
growth oxide substrates.
In this communication, we demonstrate that adding a water soluble PVA layer in-between the PMMA layer
and 2D material grown on a rigid substrate allows not only the effective transfer to arbitrary target substrates
with a high degree of freedom but also avoids the etching of the PMMA layer to obtain contamination-free high
quality 2D materials by avoiding detrimental effects related to surface contaminants. Graphene transferred by this
process exhibits excellent quality, indicated by a charge neutrality point being close to zero. Due to the elimination
of contamination from the PMMA residues, both graphene and MoS2 FETs fabricated using our transfer method
showed higher mobility and current modulation values compared to those formed using conventional PMMA
assisted transfer of the 2D materials. We describe the details of the PMMA/PVA based transfer method for 2D
materials grown on different rigid substrates, for instance, graphene on copper foil, h-BN on platinum and MoS2
on SiO2/Si substrates. The quality of the transferred 2D materials has been evaluated systematically and the ver-
satility of the transfer method is demonstrated by fabricating reliable graphene and MoS2 field effect transistors.

Results and Discussion

Figures1a,b show schematic illustrations of the PMMA/PVA assisted transfer of CVD graphene and h-BN,
respectively, onto various rigid target substrates. The detailed transfer procedure is described in the Methods
Section. There are three essential steps in the PMMA/PVA assisted transfer of graphene: first, coating PMMA/
PVA onto the graphene, second, etching of the Cu layer and third, releasing the graphene layer followed by the
transfer onto the target substrate. The crucial breakthrough in our newly developed PMMA/PVA assisted method
when compared with the conventional PMMA assisted method is in the third step. Usually, PMMA is removed
by dissolving in acetone, whereas, in the PMMA/PVA assisted method, PMMA is released in hot water at 130C.
Due to the low viscoelastic properties of PVA, PVA alone is not strong enough to support the transfer of graphene,
thus it can easily break in a form of thin film. Therefore, PMMA is still needed as an intermediate support-
ing membrane in transferring 2D materials similar to the conventional PMMA assisted 2D materials transfer
method, thus facilitating easy handling of 2D materials. On the other hand, in the PMMA/PVA assisted transfer
method, a thin PVA layer of 100nm is employed as a buffer layer between graphene and PMMA. As PVA can be
easily dissolved in hot water, PMMA can be peeled off directly, leaving graphene on the substrate. We exploited
the PVAs unique property of being easily dissolved in hot water, thus leaving no trace on the surface of the 2D
materials as demonstrated in the transfer of mechanically exfoliated 2D layers8. Therefore, with the PMMA/
PVA assisted transfer method, which has all the advantages of the conventional PMMA transfer method such as
simplicity, large area, and high degree of freedom, we can easily transfer 2D materials of different morphologies
onto arbitrary target substrates with controlled transfer orientation while minimum contamination. There are a
few studies that use a similar method for the crystal transfer912. However, it should be clearly noted that in this
dry-transfer method, the 2D material is mechanically transferred onto the top of PMMA/PVA and 2D/PMMA

Scientific Reports | 6:33096 | DOI: 10.1038/srep33096 2

www.nature.com/scientificreports/

Figure 2. Graphene transfer. Optical microscopy images of (a,c) bright field and (b,d) dark field, (e,g) AFM
images and (f,h) Raman spectra of transferred graphene on SiO2/Si substrate by copper etching method using
PMMA and PVA, respectively.

layer is released from the substrate by dissolving PVA in water and then transferred to a target substrate for
device fabrication. Therefore, the complete removal of PMMA from 2D material remains as a technical issue to
be resolved.
We have compared the quality of the graphene layers transferred onto SiO2/Si substrate by PMMA and
PMMA/PVA assisted transferring methods by using optical microscopy, atomic force microscopy (AFM), Raman
spectroscopy and electrical measurement. Figure2 shows optical microscopy (a,c) bright field and (b,d) dark
field images of graphene on SiO2/Si substrate transferred by copper etching method using PMMA and PMMA/
PVA, respectively. Compared to the PMMA method, the graphene transferred by PMMA/PVA exhibited a much
cleaner surface. The problem associated with the incomplete etching of PMMA has been well documented; ace-
tone cleaning cannot completely remove PMMA residues due to their strong Van der Waals interaction with
graphene1315. The morphological features of graphene transferred using both transfer methods were also exam-
ined using AFM (Fig.2e,g). Graphene transferred using PMMA/PVA shows a flat and clean surface after trans-
fer and no residual polymer particles are observed, in clear contrast to the graphene transferred using PMMA.
Contamination-free high quality graphene layer was obtained simply by dissolving PVA in hot water, even with-
out any additional thermal-annealing process7,15,16.
Raman spectroscopy is a quick and unambiguous method to evaluate the quality and the number of graphene
layers. Figure2f,h are representative Raman spectra of the transferred single-layer graphene on the target sub-
strate (SiO2/Si) using the PMMA and PMMA/PVA assisted transfer methods, respectively. The difference of the
D, G and 2D peaks indicates the quality and uniformity of the graphene3. The D peak, at 1,350cm1 is caused by
the presence of structural disorder, randomly distributed impurities, surface charges or doping in graphene7.
Disorder in sp2-hybridized carbon systems leads to resonance in their Raman spectra, which makes Raman
spectroscopy one of the most sensitive techniques to characterize such disorder and to evaluate the quality of
graphene1719. Here we found that the high intensity of the D peak in the case of PMMA transferred sample is
due to the presence of PMMA residues or due to wrinkles formed during the transfer process20,21. The D peak is
almost absent in the case of the PMMA/PVA transferred graphene, which suggests that the graphene sustained
physical damage and/or contamination both during the transfer processes and during PMMA/PVA removal,
thus leading to uniform and high quality transferred graphene17,18. Moreover, it has been demonstrated that the
increase in the 2D/G ratio points to a cleaner surface with lower doping level or contamination22,23. The Raman
spectra also showed that the 2D/G ratio of the graphene transferred using PMMA/PVA is greater than 2 while the
2D/G ratio of graphene transferred by PMMA is less than 2, indicating a better quality and cleaner surface for the
PMMA/PVA transferred graphene.
An additional thermal annealing was performed for graphene transferred by both PMMA and PMMA/PVA
methods in an attempt to remove polymer residues on the surface and thereby reduce the number of trapping sites
at the interface between the transferred graphene and the target SiO2/Si substrate14,16,24. The annealing was carried
out at 250C for 6h under a gas flow containing 100 sccm argon and 5% hydrogen. As shown in Supplementary
Fig. S1, even after the thermal annealing process, some PMMA residues were still observed in the AFM images
indicating the difficulty in completely removing these residues from the graphene surface. Thus, the difficulty to
completely remove residual PMMA by annealing is the main drawback that leads to a high Dirac voltage and a
large hysteretic window in FETs made from graphene transferred using PMMA14,24. The quality of the PMMA/
PVA-transferred graphene on the SiO2/Si substrate after annealing, was also evaluated by Raman spectroscopy
as shown in Supplementary Fig. S2. We observed the D band for the annealed graphene whereas we observe no
D band before annealing (Fig.2h). We therefore conclude that annealing at high temperature not only brings

Scientific Reports | 6:33096 | DOI: 10.1038/srep33096 3

www.nature.com/scientificreports/

Figure 3. Graphene FETs characteristics. (a,c) Resistivity versus the back gate voltage, (b,d) Histogram of
Dirac points, of graphene on SiO2/Si substrate transferred by copper etching method using PMMA and PVA,
respectively. Insets show the optical images of the corresponding devices.

graphene in close contact with SiO2 dielectric layer that increases the coupling between them, but also introduces
other undesirable defects, resulting in the activation of the D band. These two observations may lead to heavy
hole doping and a severe degradation of carrier modulation and reduced mobility in graphene FET devices as
reported by Cheng et al.15 Thus, taking into account the advantage of our PMMA/PVA transferring method, high
quality, large area and clean 2D materials can be easily transferred even without the need for subsequent thermal
treatment, which makes this process particularly attractive for the fabrication of various devices that need to be
fabricated at low temperature. It is obvious that the scale of the graphene transferred by PMMA/PVA can be as
large as that transferred by the conventional PMMA method, thus making it viable for large-area graphene trans-
fer for industrial applications.
To evaluate the electrical property of the transferred graphene, back-gated graphene FETs were fabricated
using graphene transferred on 100nm thick SiO2/Si substrates, with Cr/Au (20/50nm) as the source and drain
electrodes, respectively. Figure3 shows the electrical characteristics for graphene FETs in which graphene was
transferred both by the conventional PMMA and by the newly developed PMMA/PVA methods. A drastic
enhancement in the device performance was clearly observed in the PMMA/PVA transferred graphene FETs in
terms of 1) Dirac voltage (VDirac) and its distribution, 2) hysteresis, and 3) modulation of electron and hole cur-
rents. Figure3a,c represent the resistivity versus the back gate voltage and Fig.3b,d show the histograms of Dirac
voltages of the graphene FETs transferred by the copper etching method with the help of PMMA and PMMA/
PVA transfer methods while the gate voltage was scanned from 60V to 60V forward and backward. The optical
images of the corresponding graphene FETs are shown as insets in the figures. We note that, a high Dirac voltage
was observed for the PMMA transferred sample, which changed from 29 to 46V when the gate voltage was swept
forward and backward. The 17V difference between the Dirac voltages for the forward and backward gate voltage
sweep (VDirac) is a measure of the hysteresis window. The large hysteresis window and the high Dirac voltage
are mainly attributed to doping effects from both the SiO2 substrate and from the residues resulting from the
incomplete removal of PMMA15. These residues create trapping sites at the surface of the transferred graphene,
which results in a degradation of the electrical properties and reduce the reliability of graphene devices13. A
narrow distribution of the Dirac voltage is indicative of both less doping and less trapping in the transferred
graphene. For the PMMA/PVA transferred graphene FETs, a more negative shift in the Dirac voltage from 2 to
10V while the gate voltage was scanned forward and backward, and a clear reduction of the hysteresis window
(12V) were observed (Fig.3c). This result is a proof of the clean surface and low levels of doping or polymer

Scientific Reports | 6:33096 | DOI: 10.1038/srep33096 4

www.nature.com/scientificreports/

residues in the transferred graphene. An average Dirac voltage value of 4V, which is close to the ideal value for
graphene FETs (i.e. 0V), and a small hysteresis window was observed for the PMMA/PVA method, which could
be due to the inevitable doping effects originated from contamination with ionic or metal residues, to maintain
its charge neutrality14,24. It has been reported that the presence of water molecules at the graphene/SiO2 interface
enhances the molecular adsorption from the ambient environment resulting in p-doping, high Dirac voltage,
gate hysteresis, and reduced mobility in graphene FETs16,25,26. For a complete and consistent analysis, more than
100 graphene FET devices made using both PMMA and PMMA/PVA methods were evaluated; the PMMA/
PVA graphene FETs showed small average Dirac voltage (45V, N=140) while a large average Dirac voltage
(387V,N=100) was observed for the PMMA graphene FETs (Fig.3b,d). Further, a large negative shift in the
average Dirac voltage from 4V to 38V was observed for the PMMA and PMMA/PVA methods, respectively. The
effective suppression of hysteresis and low Dirac voltage in the PMMA/PVA graphene FET originates from the
absence of charge-trapping sites, i.e., polymer residues on the surface of the transferred graphene.
In addition to change in the Dirac voltage and the hysteresis window, the FETs from PMMA/PVA transferred
graphene exhibited much enhanced mobility, conductivity and improved symmetry in charge-carrier transport
as compared to the PMMA transferred graphene FETs. The mobilities of the n-type and p-type carriers of
graphene FETs are also extracted from the IdsVg transfer characteristics and are shown in Fig.3a,c. The carrier
modulation and the average electron mobility are 1,607 and 3,587cm2 V1 s1 for the PMMA and PMMA/PVA
samples, respectively and the hole mobilities are increased from 1,687 in the PMMA sample to 3,800cm2 V1 s1
for the PMMA/PVA graphene FET. Thus, the mobility for both n-type and p-type carriers is enhanced by more
than twice for the PMMA/PVA graphene. The mobility value was extracted using the Drude model, which gives
the dependence of the mobility on the carrier concentration27. The carrier mobility () and the carrier density (n)
can be calculated from the equation
1
=
nes Rsh (1)

Cox (Vg VDirac )

ns =
e (2)
where ns is the variable carrier density, which depends on the back-gate voltage; C ox is the gate oxide capacitance
(34.5nF cm2) and e is the electronic charge. The sheet resistance (Rsh) was calculated as

Rsh =
(VX 2 )
VX1 w
Id (3)
where w = 10m is the channel width, d = 10m is the distance between the voltage leads X 2 and X 1, and I is the
current flowing between the leads 1 and 2.
Apart from substrate etching methods, transfer methods based on bubbling have also been used to transfer
CVD grown 2D materials such as h-BN layer on Pt foil28 and graphene on Cu foil29,30 onto arbitrary substrates
using PMMA assisted transfer. This method avoids metal etching and the 2D material films are transferred by
direct delamination from the growth substrate. We also successfully demonstrated a PMMA/PVA assisted trans-
fer of CVD grown single layer h-BN on Pt foil onto an arbitrary substrate by using the bubbling method. The
schematic diagram of the PVA assisted transfer of CVD-grown h-BN by bubbling transfer method is shown
in Fig.1b. The detailed transfer procedure is also described in the Experimental Section. Similar to the PVA
graphene transfer process, instead of etching the copper foil in step 2, the bubbling method based on an electro-
chemical reaction was used to release the single layer of h-BN from the Pt foil28. The quality of the transferred
h-BN on 300nm SiO2/Si substrate was examined in detail by optical microscopy, SEM, AFM and Raman spec-
troscopy for both PMMA and PMMA/PVA based transfer methods for comparison. As in Fig.4, optical micros-
copy images of (a,c) bright field and (b,d) dark field, (e,g) SEM images and (f,h) AFM images of the transferred
single layer h-BN on SiO2/Si substrate by the bubbling method using PMMA and PMMA/PVA, respectively are
shown. From these results, we can conclude that, the PMMA/PVA method leads to a much cleaner sample, almost
free from polymer residue when compared to the PMMA method. The single layer h-BN was also characterized
by Raman spectroscopy but so far, we did not observe any significant difference in the Raman spectrum between
the PMMA and PMMA/PVA transferred h-BN sample (Supplementary Fig. S3). It should noted that in the case
of the PMMA/PVA transferred graphene, while no visible polymer residues were observed in the AFM images
(Fig.2g), the optical dark field image (Fig.2d) revealed the presence of some residual contamination on the
surface. These could be either ionic impurities from the metal etchant or metallic residues from the incomplete
etching of the metal. These contaminants are trapped at the interface between the transferred graphene and the
target substrate, which results in a degradation of the electrical properties and reduces the reliability of graphene
devices14,29. A clean surface free of contamination was observed in the case of PMMA/PVA transferred h-BN as
shown in Fig.4c,d,g,h. We can expect that the bubbling transfer method can therefore lead to better electrical,
optical and mechanical properties in devices based on 2D materials. In general, our PMMA/PVA method can be
applied to transfer other 2D materials grown on metal substrates such as graphene on copper foil29,30, MoS2 on
Au foil31 among others.
Finally, the PVA assisted transfer of CVD grown atomic layers of MoS2 on 300nm SiO2/Si substrate onto
arbitrary substrates was also successfully demonstrated. A schematic diagram of this process involving the etch-
ing of the oxide layer is shown in Supplementary Fig. S4. The detailed transfer procedure is described in the
Experimental Section. Similar to the PMMA/PVA graphene transfer process, the SiO2 oxide layer etching was

Scientific Reports | 6:33096 | DOI: 10.1038/srep33096 5

www.nature.com/scientificreports/

Figure 4. Single layer h-BN transfer. Optical microscopy images of (a,c) bright field and (b,d) dark field, (e,g)
SEM images, (f,h) Raman spectra and AFM images of single layer h-BN on SiO2/Si substrate transferred by
bubbling transfer method using PMMA and PMMA/PVA, respectively.

Figure 5. MoS2 transfer. Optical microscopy images of (a,c) bright field and (b,d) dark field, (e,g) AFM
images, (f) Raman spectra; (i,k) IdsVg transfer characteristics and (j) optical images of MoS2 FETs fabricated
from MoS2 on SiO2/Si substrate transferred by SiO2 substrate etching transfer method using PMMA and PVA,
respectively.

carried out instead of etching the copper foil in step 2. The quality of the transferred MoS2 on 300nm SiO2/Si
substrate was carefully examined by optical microscopy, AFM and Raman spectroscopy for both PMMA and
PMMA/PVA methods for comparison. From optical, and AFM images (Fig.5) we can conclude that the PMMA/

Scientific Reports | 6:33096 | DOI: 10.1038/srep33096 6

www.nature.com/scientificreports/

PVA method helps to transfer clean MoS2 layers onto arbitrary substrates without polymer residues. The presence
of single and multi-layer MoS2 was confirmed by Raman spectroscopy as shown in Fig.5f,i,k present the transfer
characteristics of the drain current versus gate-source voltage (IdsVg) for several n-type single layer MoS2 FETs
using MoS2 on 100nm SiO2/Si substrate transferred by the conventional PMMA and PMMA/PVA methods,
respectively are shown. These were obtained by sweeping the gate voltage continuously from 25 to +15 V with
a drain voltage of 0.1V. The transconductance (gm), the electron mobility (e), the subthreshold swing (S.S) and
the carrier concentration are estimated from the transfer characteristics of the drain current vs the gate-source
voltage (IdsVg). We also observed an enhancement in the device performance. For instance, the transconduct-
ance increased from 7.5 to 83nS, electron mobility increased from 1.1 to 12cm2 V1 s1 while subthreshold swing
reduced from 30 to 22mV dec1 for PMMA and PMMA/PVA-MoS2 FETs, respectively (Supplementary S5).
These results further confirm that the PMMA/PVA transfer method leads to materials with high quality, low lev-
els of contamination and free from polymer residues. In general, our PMMA/PVA method can be used not only
for transferring atomic layers of MoS2, but can also be applied to other 2D transition metal dichalcogenide mate-
rials grown on oxide substrates onto different types of substrates for high quality, large area, contamination-free
layers for many viable device applications.
In summary, we have developed a novel, facile transfer technique for CVD-grown 2D materials. Several 2D
materials grown on different rigid substrates such as graphene on copper foil, h-BN on platinum and MoS2 on
SiO2/Si substrates have been transferred using this method. The quality of the transferred materials is carefully
examined by optical microscopy, SEM, AFM, Raman spectroscopy and IV characteristics. Graphene and MoS2
FET devices from PMMA/PVA transferred layers showed negative shift of Dirac point and higher performance
when compared to devices fabricated using the PMMA assisted transfer process. This strategy can be extended
to several other 2D material systems grown on different types of substrates. Multilayer stacking of heterostruc-
tures of 2D materials can be reliably performed on the wafer-scale, thus paving the way to better optoelectronic,
mechanical and nanoelectronic devices.

Methods
PMMA/PVA assisted transfer of CVD-grown graphene on copper foil. The monolayer graphene
used in this study was grown on a polished copper (111) foil using CVD method, as reported previously by
Nguyen et al.32
A schematic illustration of the transfer process by PMMA/PVA assisted transfer of single-layer graphene is
shown in Fig.1a. The first step of the process involves spin-coating an aqueous PVA solution on CVD-graphene
growth substrate (graphene/copper/PET), at step1:1000rpm for 10s and step 2:3000rpm for 60s, followed by
baking at 100C for 60s using a hotplate. Aqueous PVA solution was prepared by dissolving 150mg of PVA
powder in 20ml DI water and stirring at 120C for 2h. Next, a layer of PMMA was spin-coated on top of the PVA
layer, at step1:1000rpm for 10s and step2:3000rpm for 60s. The PMMA/PVA/graphene/copper foil block was
floated on the surface of a solution of 0.3M ammonium persulfate (Aldrich, 98%) at 0C for 24h to etch the
copper foil. After the copper was etched, the PMMA/PVA/graphene block was rinsed with deionized water three
times at 0C and transferred onto a 100nm thick SiO2/Si substrate. Prior to the transfer, the SiO2/Si substrate
was treated for 5min in oxygen plasma to render the SiO2 substrate hydrophilic and ensure a better wetting of
graphene. The transferred substrate was dried under reduced pressure (~10 millitorr) for 24h and left in air for
three days. Finally, the PMMA/PVA carrier was removed by dissolving the PVA in deionized water at 130C.

PMMA/PVA assisted transfer of CVD-grown single layer h-BN on Pt substrate. The single layer
h-BN samples used in this study were grown on Pt foil by CVD method, as reported in detail by Kim et al.28
The bubbling-based transfer method was used to successfully transfer the h-BN layer grown on Pt foil onto
arbitrary substrates that allowed the Pt foil to be recycled28. Figure1b shows the schematic of the bubbling-based
transfer method, which is based on the electrolysis of water. After the growth of h-BN on Pt foil, the first layer
of PVA was coated by spin-coating an aqueous PVA solution on the h-BN/Pt foil followed by a layer of PMMA
on top of the PVA layer. Next, the structure composed of PMMA/PVA/h-BN/Pt foil was slowly dipped vertically
into a 1 M aqueous solution of NaOH at 0C. The PMMA/PVA/h-BN/Pt foil structure was used as the cathode
and a piece of bare Pt foil was used as the anode. The application of a constant current for a few min caused the
PMMA/PVA/h-BN layer to detach from the Pt foil due to the formation of bubbles upon H2 evolution. After the
completion of the transfer process, the PMMA/PVA/h-BN layer was rinsed with deionized water at 0C three
times to remove any remaining NaOH and transferred onto the target substrate. It should be noted that the Pt
foil used in the growth of h-BN could be recycled since it was not consumed during the process. The transferred
substrate is dried under reduced pressure (~10 millitorr) for 24h and left in air for three days. Finally, the PMMA/
PVA carrier is removed by dissolving the PVA in deionized water at 130C.

PMMA/PVA assisted transfer of CVD-grown single layer MoS2 on SiO2/Si substrate. The atomic
layers of MoS2 used in this study were grown on 300nm SiO2/Si substrate by CVD, as reported in detail by Yu et al.33
After the growth of MoS2 on SiO2/Si substrate, the first layer of PVA was coated by spin-coating an aqueous
PVA solution on the MoS2/SiO2/Si substrate. The substrate was baked at 100C for 1min on a hotplate. This is
followed by coating a second layer of PMMA on top of the PVA layer. The PMMA/PVA/MoS2/SiO2/Si substrate
block was floated on the surface of 1 M Potassium hydroxide solution (Aldrich, 98%) at 0C for 48h to etch
the SiO2. After the etching, the PMMA/PVA/MoS2 block was rinsed three times with deionized water at 0C
and transferred onto a 100nm thick SiO2/Si substrate. Prior to the transfer, the SiO2/Si substrates were treated
for 5min in oxygen plasma to render them hydrophilic to ensure better wetting of graphene. The transferred

Scientific Reports | 6:33096 | DOI: 10.1038/srep33096 7

www.nature.com/scientificreports/

substrate was then dried under reduced pressure (~10 millitorr) for 24h and left in air for three days. Finally, the
PMMA/PVA carrier was removed by dissolving PVA in deionized water at 130C.

Fabrication and measurement of Graphene and MoS2 field effect transistors. Graphene and
MoS2 were transferred onto a silicon substrate using both PMMA and PMMA/PVA methods. A heavily doped
p-type Si substrate (0.005cm) was employed as the back gate with a 100-nm-thick, thermally oxidized SiO2 top
layer as the gate oxide layer. Multiple electrodes were patterned on the 2D materials by a conventional photoli-
thography process. Subsequently, contact electrodes to form ohmic contacts were deposited by electron-beam
evaporation as follows: a 20-nm-thick chromium layer was first evaporated, followed by a 50-nm-thick Au layer.
The electrical characteristics of the devices were measured in vacuum using a probe station with a Keithley
SCS-4200 system.

References
1. Caldwell, J. D. et al. Technique for the Dry Transfer of Epitaxial Graphene onto Arbitrary Substrates. ACS Nano 4, 11081114 (2010).
2. Lee, Y. et al. Wafer-Scale Synthesis and Transfer of Graphene Films. Nano Lett. 10, 490493 (2010).
3. Chen. X-D. et al. High-quality and efficient transfer of large-area graphene films onto different substrates. Carbon 56, 271278
(2013).
4. Kang, J. et al. Efficient Transfer of Large-Area Graphene Films onto Rigid Substrates by Hot Pressing. ACS Nano 6, 53605365
(2012).
5. Yoon, T. et al. Direct Measurement of Adhesion Energy of Monolayer Graphene As-Grown on Copper and Its Application to
Renewable Transfer Process. Nano Lett. 12, 1448-1452 (2012).
6. Cheol Shin, W. et al. Doping suppression and mobility enhancement of graphene transistors fabricated using an adhesion promoting
dry transfer process. Appl. Phys. Lett. 103, 243504 (2013).
7. Yang, S. Y. et al. Metal-Etching-Free Direct Delamination and Transfer of Single-Layer Graphene with a High Degree of Freedom.
Small 11, 175181 (2015).
8. Banszerus, L. et al. Ultrahigh-mobility graphene devices from chemical vapor deposition on reusable copper. Sci. Adv. 1, (2015).
9. Leon, J. A. et al. Transferring few-layer graphene sheets on hexagonal boron nitride substrates for fabrication of graphene devices.
Graphene 3, 2535 (2014).
10. Henrique Guimaraes, R., Jos Carlos Viana, G. & Eunzio, A. T. d. S. Transfer of an exfoliated monolayer graphene flake onto an
optical fiber end face for erbium-doped fiber laser mode-locking. 2D Mater. 2, 031001 (2015).
11. Dean, C. R. et al. Boron nitride substrates for high-quality graphene electronics. Nat. Nano. 5, 722726, (2010).
12. Garcia, A. G. F. et al. Effective Cleaning of Hexagonal Boron Nitride for Graphene Devices. Nano Letters 12, 44494454, (2012).
13. Mun, J. H. & Cho, B. J. Synthesis of Monolayer Graphene Having a Negligible Amount of Wrinkles by Stress Relaxation. Nano Lett.
13, 24962499 (2013).
14. Liang, X. et al. Toward Clean and Crackless Transfer of Graphene. ACS Nano 5, 91449153 (2011).
15. Cheng, Z. et al. Toward Intrinsic Graphene Surfaces: A Systematic Study on Thermal Annealing and Wet-Chemical Treatment of
SiO2-Supported Graphene Devices. Nano Lett. 11, 767771 (2011).
16. Chan, J. et al. Reducing Extrinsic Performance-Limiting Factors in Graphene Grown by Chemical Vapor Deposition. ACS Nano 6,
32243229 (2012).
17. Canado, L. G. et al. Quantifying Defects in Graphene via Raman Spectroscopy at Different Excitation Energies. Nano Lett. 11,
31903196 (2011).
18. Tao, L. et al. Uniform Wafer-Scale Chemical Vapor Deposition of Graphene on Evaporated Cu (111) Film with Quality Comparable
to Exfoliated Monolayer. J. Phys. Chem. C 116, 2406824074 (2012).
19. Wei, D. et al. Synthesis of N-Doped Graphene by Chemical Vapor Deposition and Its Electrical Properties. Nano Lett. 9, 17521758
(2009).
20. Bae, S. et al. Roll-to-roll production of 30-inch graphene films for transparent electrodes. Nat. Nanotechnol. 5, 574578 (2010).
21. Reina, A. et al. Transferring and Identification of Single- and Few-Layer Graphene on Arbitrary Substrates. J. Phys. Chem. C 112,
1774117744 (2008).
22. REN, Y. et al. An Improved Method For Transferring Graphene Grown By Chemical Vapor Deposition. Nano 07, 1150001 (2012).
23. Casiraghi, C., Pisana, S., Novoselov, K. S., Geim, A. K. & Ferrari, A. C. Raman fingerprint of charged impurities in grapheme. Appl.
Phys. Lett. 91, 233108 (2007).
24. Gao, L. et al. Repeated growth and bubbling transfer of graphene with millimetre-size single-crystal grains using platinum. Nat.
Commun. 3, 699 (2012).
25. Nagashio, K., Yamashita, T., Nishimura, T., Kita, K. & Toriumi, A. Electrical transport properties of graphene on SiO2 with specific
surface structures. J. Appl. Phys. 110, 024513 (2011).
26. Ni, Z. H. et al. The effect of vacuum annealing on graphene. J. Raman Spectrosc. 41, 479483 (2010).
27. Lee, J.-H. et al. Wafer-Scale Growth of Single-Crystal Monolayer Graphene on Reusable Hydrogen-Terminated Germanium. Science
344, 286289 (2014).
28. Kim, G. et al. Growth of High-Crystalline, Single-Layer Hexagonal Boron Nitride on Recyclable Platinum Foil. Nano Lett. 13,
18341839 (2013).
29. de la Rosa, C. J. L. et al. Frame assisted H2O electrolysis induced H2 bubbling transfer of large area graphene grown by chemical
vapor deposition on Cu. Appl. Phys. Lett. 102, 022101 (2013).
30. Gao, L. et al. Repeated growth and bubbling transfer of graphene with millimetre-size single-crystal grains using platinum. Nat.
Commun. 3, 699 (2012).
31. Shi, J. et al. Controllable Growth and Transfer of Monolayer MoS2 on Au Foils and Its Potential Application in Hydrogen Evolution
Reaction. ACS Nano 8, 1019610204 (2014).
32. Nguyen, V. L. et al. Seamless Stitching of Graphene Domains on Polished Copper (111) Foil. Adv. Mater. 27, 13761382 (2015).
33. Yu, Y. et al. Controlled Scalable Synthesis of Uniform, High-Quality Monolayer and Few-layer MoS2 Films. Sci. Rep. 3, 1866 (2013).

Acknowledgements
This work was supported by the Power Generation and Electricity Delivery Core Technology Program of the
Korea Institute of Energy Technology Evaluation and Planning (KETEP) that granted financial resources
from the Ministry of Trade, Industry and Energy, Republic of Korea (no. 20131020400830). D.J.K. would also
like to acknowledge the financial support provided by the National Research Foundation of Korea (NRF-
2014R1A2A1A11052965).

Scientific Reports | 6:33096 | DOI: 10.1038/srep33096 8

www.nature.com/scientificreports/

Author Contributions
H.V.N. and D.J.K. conceived and designed the project. H.V.N., Y.Q. and S.K.H. performed the experiments and
analysed the data. H.V.N. and D.J.K. wrote the manuscript. D.J.K. supervised the project. All authors discussed
the results and commented on the manuscript.

Additional Information
Supplementary information accompanies this paper at http://www.nature.com/srep
Competing financial interests: The authors declare no competing financial interests.
How to cite this article: Van Ngoc, H. et al. PMMA-Etching-Free Transfer of Wafer-scale Chemical Vapor
Deposition Two-dimensional Atomic Crystal by a Water Soluble Polyvinyl Alcohol Polymer Method. Sci. Rep. 6,
33096; doi: 10.1038/srep33096 (2016).
This work is licensed under a Creative Commons Attribution 4.0 International License. The images
or other third party material in this article are included in the articles Creative Commons license,
unless indicated otherwise in the credit line; if the material is not included under the Creative Commons license,
users will need to obtain permission from the license holder to reproduce the material. To view a copy of this
license, visit http://creativecommons.org/licenses/by/4.0/

The Author(s) 2016

Scientific Reports | 6:33096 | DOI: 10.1038/srep33096 9